Web-based screening of neurological diseases using pose estimator

2022 Fiscal Year Final Research Report

• Project/Area Number: 21K20891
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Multi-year Fund
• Review Section: 0902:General internal medicine and related fields
• Research Institution: The University of Tokyo
• Principal Investigator: Sato Kenichiro 東京大学, 大学院医学系研究科(医学部), 助教 (10908495)
• Project Period (FY): 2021-08-30 – 2023-03-31
• Keywords: 歩行解析 / 自動解析 / 姿勢推定 / 深層学習 / 時計描画テスト / 認知機能低下 / スクリーニング / AI

Outline of Final Research Achievements

We aimed to detect neurodegenerative diseases including Alzheimer's disease (AD) or Parkinson's disease (PD) by web-based applications available in smartphones, by attempting to develope two different methods: one is applying pose-estimating to 2D video movies recording gait, and another is deep learning-based prediction of cognitive decline from Clock-Drawing Test pictures. We started to take data of gait movies from preclinical AD participants. We also built deep learning prediction models to identify those with probable dementia or with executive dysfunction.

Free Research Field

Neurology

Academic Significance and Societal Importance of the Research Achievements

本研究の発展により、web上(含スマートフォン)での早期検出を目的としたアプリケーションを用意することができ、より幅広い人に利用してもらうことが可能になるため、従って神経疾患の早期発見がより多くの人で可能になるのではないかと期待できる。